© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 8684
    Prev
    Next
    INDEX
    Explanations

    only during, confess feelings, matters within

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on numeric tokens (especially decimal or large-digit figures) indicating quantities or statistics.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ട്ടുണ്ട്
    0.71
     obliter
    0.71
    깐
    0.71
     аль
    0.70
    йо
    0.69
    sser
    0.68
     inactiv
    0.68
     брю
    0.68
     буде
    0.67
    لون
    0.67
    POSITIVE LOGITS
    erade
    0.88
     oryg
    0.85
     pessoais
    0.79
     iż
    0.78
     capacidades
    0.78
     "../../
    0.76
     clientes
    0.75
     හේ
    0.75
     sú
    0.75
     hợp
    0.74
    Activations Density 0.000%

    No Known Activations